Reviews/Osteoporosis Fracture Prevention/Denosumab versus bisphosphonate bone density
Denosumab Therapy · ostfrac_07 · v2
Borderline: within 4 points of the Established/Likely boundary, so a re-grade could land either side.
Head-to-head studies that put denosumab directly against bisphosphonates consistently find denosumab produces larger gains in bone density, and none of them find the reverse. The strongest designs are a randomized trial [1] and a pooled analysis of four randomized trials [19], both of which show greater density gains with denosumab at every site measured. Two smaller studies [3][5] point the same way but only reach statistical significance at a single site each. A very large real-world study [17] adds support by showing denosumab lowers fracture risk more than a bisphosphonate. The panel settled where it did because the direction is the same across independent studies with nothing contradicting it, while the gains are modest in size and most of the cleanest data come from women switching off a prior bisphosphonate rather than starting treatment fresh.
Every direct comparison in this corpus favours denosumab for bone mineral density and none favours bisphosphonates. The randomized open-label trial [1] reports greater gains with denosumab at total hip (2.3% vs 1.1%), femoral neck (1.7% vs 0.7%) and lumbar spine (4.1% vs 2.0%; P<.001 at all sites), and the pooled analysis of four RCTs [19] (N=2850) finds significantly greater gains (p<0.001) at each skeletal site with between-group differences of 0.6 to 2.0%. Two further comparisons agree in direction but are site-restricted: [3] significant at lumbar spine only and [5] showing a femoral-neck difference of 4.59% (95% CI 0.93% to 8.25%; P=.017) but not at lumbar spine or total hip. The comparative-effectiveness cohort [17] (denosumab n=89115 vs alendronate n=389536) corroborates downstream, reporting fracture-risk reductions of 39% for major osteoporotic, 36% hip and 43% nonvertebral. The contested point: The division is over scope and uniformity, not direction: the two cleanest datasets [1][19] are in women previously treated with bisphosphonates, so the evidence speaks to a switch scenario more than to treatment-naive comparison. The panel also divides on how much weight to give results that reach significance at only one site each in [3] and [5], and on whether a 0.6 to 2.0% difference [19] is clinically meaningful.
All 23 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Osteoporosis Fracture Prevention, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The cleanest head-to-head datasets [1] and [19] are in postmenopausal women previously treated with bisphosphonates (transition populations), so the unqualified claim over-generalises from a specific switch scenario.
Two of four direct comparisons show site-restricted superiority: [3] significant only at lumbar spine (TH/FN not), [5] only at femoral neck (LS/TH not); 'more BMD' is therefore not uniform across skeletal sites.
Where it remains uncertain. The strongest head-to-head data [1][19] come from bisphosphonate-pretreated (transition) populations, and two smaller studies [3][5] reach significance at only one skeletal site each, so superiority is neither treatment-naive nor uniform across sites.
Aleatory
In the smaller comparisons [3][5] (N=140, N=120), between-group differences reach significance at only one site each, consistent with sampling noise at the other sites.
Epistemic
The strongest datasets [1][19] enrol women previously on bisphosphonates; the corpus contains no direct treatment-naive head-to-head comparison.
Inferential
The claim is about BMD, but [17] states head-to-head fracture RCTs are lacking, so inferring patient-important benefit from density gains is not supported by this corpus.
The load-bearing papers from a corpus of 23, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Direct RCT evidence of greater denosumab BMD gain at all three sites vs ibandronate.
“At month 12, significantly greater BMD gains from baseline were observed with denosumab compared with ibandronate at the total hip (2.3% compared with 1.1%), femoral neck (1.7% compared with 0.7%), and lumbar spine (4.1% compared with 2.0%; treatment difference P<.001 at all sites).”
Panel auditInternal validity: high · scope of inference: Postmenopausal women with low BMD previously treated with a bisphosphonate, over 12 months.Key limitations: open-label · prior bisphosphonate exposure · only ibandronate as comparator
Consistent direction but superiority confined to lumbar spine in a post-teriparatide cohort.
“LS, but not TH or FN, BMD gain was significantly greater in the denosumab group than in the bisphosphonates group at 12 months.”
Panel auditInternal validity: medium · scope of inference: Postmenopausal women with severe osteoporosis after teriparatide, over 12 months.Key limitations: retrospective · non-randomized · benefit only at lumbar spine · post-teriparatide context
Consistent direction but superiority confined to femoral neck in a T2DM population.
“After 1-year of treatment, DEN was more effective than ZOL at increasing femoral neck BMD (least-squares mean [LS mean] difference 4.59% [95% confidence interval [CI]: 0.93% to 8.25%]; P = .017), but not at BMD in the lumbar spine or total hip.”
Panel auditInternal validity: medium · scope of inference: Osteoporotic women aged 50-80 with type 2 diabetes, over 1 year.Key limitations: non-randomized · preference allocation · diabetes-specific population · benefit only at femoral neck
3 of 23 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Multiple independent head-to-head primary studies consistently show greater BMD gain with denosumab, led by an RCT and a large pooled RCT analysis, with modest magnitude and transition-population scope keeping confidence at the lower bound of Established.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Treatment-naive head-to-head comparison. The cleanest BMD data [1][19] are confined to bisphosphonate-pretreated women, leaving the unqualified claim unsupported for first-line initiation. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A randomized trial of denosumab versus a bisphosphonate in treatment-naive osteoporosis patients with BMD change at all sites as the endpoint.
Open question: Does denosumab increase BMD more than a bisphosphonate in treatment-naive patients, not only after switching? Why it is still open: The strongest comparisons [1][19] enrolled only bisphosphonate-pretreated women, and the corpus has no naive head-to-head.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19